International Journal of Humanoid Robotics

This study presents an enhanced framework for evaluating an assistive effect generated by a passive assistive device using a humanoid robot. The humanoid robotic experiments can evaluate wearable devices by measuring the joint torque, which cannot be measured directly from the human body. In this paper, we introduce an “assistive torque estimation map” as an efficient means for estimating the supportive torque within the range of motions by interpolating the measured joint torques and joint angles of the robot. This map aims to estimate the supportive torques for complex motions without conducting humanoid experiments or human-subject experiments with these motions. We generated an estimation map for an actual assistive suit that decreases the load on the lumbar region and we verified the validity of the proposed method by experimentation. In addition, the geometric simulation model of the assistive suit was validated based on the proposed experiments by using the humanoid robot HRP-4. The proposed framework is expected to lead to an efficient design of such assistive devices so that fewer human-subject experiments need to be conducted.

This paper presents a control method of the torso for dynamic walking of biped robots. Specifically, we want to save the energy and improve the walking stability by the planning and control of torso orientation at landing. First, the impact process of leg exchange is formulated using a simplified model. Based on this, the influence of the torso orientation at landing on walking performance is investigated. Second, a control method of torso orientation, which is an under actuated control, is proposed to regulate the torso orientation in the single support phase (SSP). Third, the control module of the torso is integrated into the previously established control frame of 3D biped walking to implement the control objective of a 3D humanoid robot. The results of a number of simulations show the feasibility of the proposed method, and also explore the relations between the speed, stability and energy consumption with the landing orientation of the torso.

In order to create effective storytelling agents three fundamental questions must be answered: first, is a physically embodied agent preferable to a virtual agent or a voice-only narration? Second, does a human voice have an advantage over a synthesized voice? Third, how should the emotional trajectory of the different characters in a story be related to a storyteller’s facial expressions during storytelling time, and how does this correlate with the apparent emotions on the faces of the listeners? The results of two specially designed studies indicate that the physically embodied robot produces more narrative attention to the listener as compared to a virtual embodiment, that a human voice is preferable over the current state of the art of text-to-speech, and that there is a complex yet interesting relation between the emotion lines of the story, the facial expressions of the narrating agent, and the emotions of the listener, and that the empathizing of the listener is evident through its facial expressions. This work constitutes an important step towards emotional storytelling robots that can observe their listeners and adapt their style in order to maximize their effectiveness.

This paper presents the balancing control performance of a mobile manipulator built in the laboratory as a service robot called Korean robot worker (KOBOKER). The robot is designed and implemented with two wheels as a mobile base and two arms with six degrees-of-freedom each. Kinematics and dynamics of the robot are analyzed. For the balancing control performance, two wheels are controlled independently by the time-delayed control method based on the inertia model of the robot. The acceleration information obtained directly from the sensor is used for the modified disturbance observer structure called an acceleration-based disturbance observer (AbDOB). Experimental studies of the balancing control of the robot are conducted to compare the control performances by both a PID control method and an AbDOB.

This paper describes research towards understanding haptic communication during planar object manipulation. In particular, a classification algorithm that classifies four stages of manipulation of a planar object is described. This research was performed as a part of a broader research project which has the goal of developing a user-friendly communication interface for an elderly-assistive robot. The manipulation of planar object was studied in detail as it happened very frequently during user study involving a caregiver helping an elderly person with the activities of daily living. For observing human haptic interaction, a sensory glove was developed. Further data collection was conducted in the laboratory setting and data was analyzed using various machine learning techniques. Based on this analysis, decision rules were derived that give insight into human-to-human collaborative manipulation of planar objects and successfully identified several classes of manipulative actions. The developed decision tree-based algorithm was then tested on the data of a user study that involved a caregiver assisting an elderly person in the activities of daily living. The developed algorithm also successfully classifies manipulation actions in real-time. This information is particularly interesting as it does not depend on any particular sensor and thus can be used by other researchers to further study haptic communication.